乳腺癌腋窝淋巴结B超检查特异性回顾性分析

New Evaluation of Ultrasonography for Axillary Lymph Node Metastasis in Breast Cancer

目的探讨各辅助检查方法对乳腺癌腋窝淋巴结转移的诊断价值。方法对天津医科大学附属肿瘤医院2009年3月至2009年6月253例经手术病理证实乳腺癌病人术前B超与钼靶、MRI、PET-CT资料进行回顾性分析,比较灵敏度、特异度、阳性及阴性预测值和准确性。结果 B超(253例)灵敏度,特异度,阳性及阴性预测值,准确性为70.6%,87.4%,84.8%,75.0%和79.1%;钼靶(220例)为14.6%,100%,100%,53.9%和57.3%;MRI(27例)为50.0%,100%,100%,71.4%,77.8%;PET-CT(23例)为90.0%,92.3%,90.0%,92.3%,91.3%。B超与病理对照的Kappa值为0.581,与病理的一致性一般;高年资组B超医师的灵敏度、特异度、准确性为与低年资组比较差异有统计学意义(P<0.05)。结论判断乳腺癌腋窝淋巴结转移状况B超优于其他检查,而且超声检查者的经验影响诊断结果。

Objective To reduce the postoperative complications of breast cancer and evaluate the value of uhrasonography in detecting ALN metastasis compared with mammography, MRI and PET-CT in patients with breast cancer. Methods 253 breast cancer patients who undergone ultrasonography were studied. The results of the histopathology after surgery were used as diagnostic golden standard. The findings of their sensitivity, specificity, positive predictive value, negative predictive value and accuracy in diagnosing axillary metastases lymph node of breast cancer were compared. Results The sensitivity, specificity, positive predictive value, negative predictive value and accuracy obtained by ultrasound were 70.6%, 87.4%, 84.8%, 75.0% and 79.1%. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy obtained by mammography were 14.6%, 100%, 100%, 53.9% and 57.3%. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy obtained by MRI were 50.0%, 100%, 100%, 71.4%, 77.8%. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy obtained by PET-CT were 90.0%, 92.3%, 90.0%, 92.3%, 91.3%. The Kappa value was 0.729. The sensitivity, specificity and accuracy of experienced doctors and un-experienced doctors were 69.2%, 87.1% and 78.1% ; 62.5 %, 88.2% and 75.8% respectively （P〈0.05）. Conclusion Our study indicates that uhrasonography is better than other examinations for predicting axillary node status, moreover the experiences of doctors affect the predicting results.